CN107025383A - Advances in protein structure prediction based on multi-objective particle swarm optimization - Google Patents
Advances in protein structure prediction based on multi-objective particle swarm optimization Download PDFInfo
- Publication number
- CN107025383A CN107025383A CN201710244181.6A CN201710244181A CN107025383A CN 107025383 A CN107025383 A CN 107025383A CN 201710244181 A CN201710244181 A CN 201710244181A CN 107025383 A CN107025383 A CN 107025383A
- Authority
- CN
- China
- Prior art keywords
- particle
- energy function
- function value
- energy
- formula
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
- G16B15/00—ICT specially adapted for analysing two-dimensional or three-dimensional molecular structures, e.g. structural or functional relations or structure alignment
-
- G—PHYSICS
- G16—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
- G16B—BIOINFORMATICS, i.e. INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR GENETIC OR PROTEIN-RELATED DATA PROCESSING IN COMPUTATIONAL MOLECULAR BIOLOGY
- G16B20/00—ICT specially adapted for functional genomics or proteomics, e.g. genotype-phenotype associations
Landscapes
- Life Sciences & Earth Sciences (AREA)
- Physics & Mathematics (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Engineering & Computer Science (AREA)
- Health & Medical Sciences (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Chemical & Material Sciences (AREA)
- Biophysics (AREA)
- Theoretical Computer Science (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Biotechnology (AREA)
- Evolutionary Biology (AREA)
- Analytical Chemistry (AREA)
- Proteomics, Peptides & Aminoacids (AREA)
- Molecular Biology (AREA)
- Genetics & Genomics (AREA)
- Crystallography & Structural Chemistry (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The present invention provides a kind of Advances in protein structure prediction based on multi-objective particle swarm optimization, including step：S1：Selection needs n optimized different initial configurations to represent that n is the natural number more than zero in the form of conformation is represented as n intended particle, and by the position coordinates of each intended particle in a protein sequence to be predicted；S2：Parameter setting is carried out to a more new formula；S3：Three target energy formula are iterated according to the more new formula, and obtain a more excellent disaggregation；S4：Handle the optimal solution set and obtain an optimal solution, and using the optimal solution as the intended particle predicted position.A kind of Advances in protein structure prediction based on multi-objective particle swarm optimization of the present invention, using multi-objective particle swarm method, the problem of for field of force function or inaccurate energy function, carries out multiple target search using three energy functions, has the advantages that validity is high and forecasting accuracy is high.
Description
Technical field
The present invention relates to biomedical sector, more particularly to a kind of protein structure based on multi-objective particle swarm optimization are pre-
Survey method.
Background technology
Protein structure refers to tertiary structure, that is, refers to a polypeptide chain on the basis of secondary structure or domain,
Further coiling, folding, fixed formed particular space structure is maintained by secondary key.Protein structure is studied, is helped
In the effect for understanding protein, understand how protein exercises its biological function, understanding protein-protein (or other points
Son) between interaction, this, either for biology still for medical science and pharmacy, is all very important.It is logical at present
Cross experimental method and determine that the process of protein structure is still extremely complex, cost is higher, it is necessary to expend substantial amounts of cost and time.
Data acquisition technology develops rapidly the biological data for generating magnanimity in recent years, it is therefore desirable to develops computational method and comes pre-
Protein structure is surveyed, biological data is made up and produces ability and understand the gap between speed.
Protein structure prediction optimization refers to the not high pre- geodesic structure of initial, precision passing through suitable chess game optimization side
Method obtains the higher protein structure of precision.The method of current protein structure optimization can be largely classified into two classes：Based on molecule
Kinetic simulation fits Monte-carlo Simulation Method.The general principle of optimization method based on molecular dynamics simulation is to apply position
Intramolecular and intermolecular interaction described by function, according to rail of the newtonian motion Mechanics Calculation protein in phase space
Mark, searches out the field of force most low state.The accuracy of this method depends on the accuracy of field of force function and the length of simulated time
It is short.Basic thought based on Monte Carlo simulation is the search in energy space using Boltzmann distribution function realization, equally
Also rely on the accuracy of energy function.
In protein structure prediction optimization, whether it is dependent on based on molecular dynamics simulation or Monte Carlo simulation
In the accuracy of field of force function or energy function, but we lack accurate energy function effectively to search at present.For
For protein, the field of force of molecule includes atomic charge, intermolecular effect gesture etc. comprising many parameters, because complexity is led
Cause many position functions and energy function that presently, there are are no can be widely used.In protein structure prediction,
We search for the structure of minimum energy state by methods such as molecular dynamics simulation or Monte Carlo simulations, and minimum energy
The structure of state be typically close to prototype structure, if so description protein structure energy energy function it is inaccurate, that
Just it is difficult to search the state close to prototype structure.
The content of the invention
For above-mentioned deficiency of the prior art, the present invention provides a kind of protein knot based on multi-objective particle swarm optimization
Structure Forecasting Methodology, using multi-objective particle swarm method, using two energy the problem of for field of force function or inaccurate energy function
Flow function carries out multiple target search, has the advantages that validity is high and forecasting accuracy is high.
To achieve these goals, the present invention provides a kind of protein structure prediction side based on multi-objective particle swarm optimization
Method, including step：
S1：Selection needs the n different initial configurations optimized as n target grain in a protein sequence to be predicted
Son, and the position coordinates of each intended particle is represented in the form of conformation is represented, n is the natural number more than zero；
S2：Parameter setting is carried out to a more new formula；
S3：Three target energy formula are iterated according to the more new formula, and obtain a more excellent disaggregation；
S4：Handle the optimal solution set and obtain an optimal solution, and using the optimal solution as the intended particle prediction
Position.
Preferably, the more new formula includes formula (1) and formula (2)：
Wherein, w is inertia coeffeicent, and k is iterations, c1For cognitive coefficient, c2For coefficient of association；For renewal speed,
The renewal speed represents that the i-th particle compares the speed for changing structure in upper once iterative process in kth time iterative process；
Rand is the random number between 0~1；For history optimum structure；For global optimum's structure,For i-th
The position coordinates of the son in kth time iterative process.
Preferably, in the S2 steps, the renewal speed for initializing each intended particle is 0, and to one most
Big iterations, the inertia coeffeicent, the cognitive coefficient and the coefficient of association are configured.
Preferably, the target energy formula includes Rosetta energy functions, QUARK energy functions and CHARMM energy
Function.
Preferably, the S3 steps further comprise step：
S31：Be utilized respectively the target energy formula calculate obtain each primary one first energy function value,
One second energy function value and one the 3rd energy function value；
S32：It is determined that and select non-dominant particle in each intended particle, and the non-dominant particle is added one more
Excellent disaggregation；The first energy function value corresponding to the non-dominant particle is the minimum value in each first energy function value,
The second energy function value corresponding to the non-dominant particle is the minimum value in each second energy function value, the non-branch
It is the minimum value in each 3rd energy function value, and the first energy letter with the 3rd energy function value corresponding to particle
Numerical value is that unique minimum value or second energy function value are each second energy in each first energy function value
Unique minimum value or the 3rd energy function value are unique minimum value in each 3rd energy function value in functional value；
S33：The renewal speed according to corresponding to the formula (1) updates each intended particle；
S34：The position coordinates of each intended particle is updated according to the formula (2)；
S35：It is utilized respectively target energy formula described in two and calculates the first energy described in the one of each intended particle of acquisition
Functional value, the 3rd energy function value described in the second energy function value and one described in one；
S36：It is determined that and select non-dominant particle in each presently described intended particle, and the non-dominant particle is added into institute
State more excellent disaggregation；
S37：Judge whether iterations reaches the maximum iteration；Such as no, return to step S33；In this way, after continuation
Continuous step.
Preferably, the S4 further comprises step：
S41：Build a utility function, U=λ1f1+λ2f2+λ3f3, wherein f1For first energy function value, f2To be described
Second energy function value, f3For the 3rd energy function value, λ1For the first weight, λ2For the second weight, λ3For the 3rd weight,
λ1、λ2And λ3Span be respectively [0,1] and λ1+λ2+λ3=1；
S42：To λ1、λ2N times stochastical sampling is carried out, the phase corresponding to each non-dominant particle of the more excellent solution concentration is calculated
Hope effectiveness E (U)；
Choose numerical value it is maximum one described in corresponding to expected utility one described in non-dominant particle be used as affiliated optimal solution.
The present invention makes it have following beneficial effect as a result of above technical scheme：
Three target energy formula are iterated, by the optimization method of multiple target, effectively improves and uses single energy
The problem of function is inaccurate.By modified particle swarm optiziation, can more effectively it search for.Meanwhile, this method improves entirety
The degree of accuracy of detection.
Brief description of the drawings
Fig. 1 is the flow with the Advances in protein structure prediction based on multi-objective particle swarm optimization of the embodiment of the present invention
Figure.
Embodiment
Below according to accompanying drawing 1, presently preferred embodiments of the present invention is provided, and is described in detail, makes to be better understood when this
Function, the feature of invention.
Referring to Fig. 1, a kind of Advances in protein structure prediction based on multi-objective particle swarm optimization of the embodiment of the present invention,
Including step：
S1：Selection needs the n different initial configurations optimized as n target grain in a protein sequence to be predicted
Son, and the position coordinates of each intended particle is represented in the form of conformation is represented, specific manifestation form isφ andDihedral angle of the intended particle in protein structure is respectively corresponded to, n is the natural number more than zero.
S2：Parameter setting is carried out to a more new formula.
Wherein, in S2 steps, the renewal speed for initializing each intended particle is 0, and to a maximum iteration, inertia
Coefficient, cognitive coefficient and coefficient of association are configured.
S3：Three target energy formula are iterated according to more new formula, and obtain a more excellent disaggregation；
Target energy formula is used in the energy function for being used for describing protein structure existing at present, the present embodiment, mesh
Mark energy theorem and use Rosetta energy functions, QUARK energy functions and CHARMM energy functions.
Wherein, S3 steps further comprise step：
S31：It is utilized respectively three target energy formula and calculates one first energy function value for obtaining each primary, one second
Energy function value and one the 3rd energy function value；
S32：It is determined that and select non-dominant particle in each intended particle, and non-dominant particle is added into a more excellent disaggregation；
The first energy function value corresponding to non-dominant particle is the minimum value in each first energy function value, corresponding to non-dominant particle
The second energy function value be each second energy function value in minimum value, and the first energy function value be each first energy function
Unique minimum value or the second energy function value are unique minimum value in each second energy function value in value；
S33：Renewal speed according to corresponding to formula (1) updates each intended particle：
Wherein, w is inertia coeffeicent, and k is iterations, c1For cognitive coefficient, c2For coefficient of association；For renewal speed,
Renewal speed represents that the i-th particle compares the speed for changing structure in upper once iterative process in kth time iterative process；Rand is
Random number between 0~1；For history optimum structure；For global optimum's structure；Exist for the i-th particle
Position coordinates in -1 iterative process of kth.
S34：The position coordinates of each intended particle is updated according to formula (2)：
For position coordinates of i-th particle in kth time iterative process.
S35：It is utilized respectively three target energy formula and calculates one first energy function value for obtaining each intended particle and one the
Two energy function values and one the 3rd energy function value；
S36：It is determined that and select non-dominant particle in each current goal particle, and the non-dominant particle is added into more excellent solution
Collection；
S37：Judge whether iterations reaches maximum iteration；Such as no, return to step S33；In this way, follow-up step is continued
Suddenly.
S4：Handle optimal solution set obtain an optimal solution, and using optimal solution as intended particle predicted position.
Wherein, S4 further comprises step：
S41：With one effectiveness of the linear of the first energy function value, the second energy function value and the 3rd energy function value and structure
Function, U=λ1f1+λ2f2+λ3f3, wherein f1For the first energy function value, f2For the second energy function value, f3For the 3rd energy letter
Numerical value, λ1For the first weight, λ2For the second weight, λ3For the 3rd weight, λ1、λ2And λ3Span be respectively [0,1] and λ1+
λ2+λ3=1；
S42：To λ1、λ2N times stochastical sampling is carried out, more excellent solution is calculated and concentrates the expectation corresponding to each non-dominant particle to imitate
With E (U)；In the present embodiment, N is more than or equal to 10000；
Choose the non-dominant particle described in the one of numerical value maximum corresponding to expected utility and be used as affiliated optimal solution.
The method of the present invention can be realized by a kind of protein structure prediction system based on multi-objective particle swarm optimization, be somebody's turn to do
System includes：One initial configuration representation module, a multi-objective particle swarm optimization module and a follow-up decision module, initial configuration table
Show that module is connected with multi-objective particle swarm optimization module, multi-objective particle swarm optimization module is connected with follow-up decision module.
Wherein, initial configuration representation module is used for the general pdb of protein structure (protein three-dimensional structure data texts
Part) conversion is represented for ease of the vectorial X of calculating, namely selection needs optimize in a protein sequence to be predicted n are not
With initial configuration as n intended particle, and the position coordinates of each intended particle is represented in the form of conformation is represented.
Multi-objective particle swarm optimization module is used to carry out parameter setting to default one more new formula, according to more new formula pair
Three target energy formula are iterated, and obtain a more excellent disaggregation；
Follow-up decision module be used for handle optimal solution set obtain an optimal solution, and using optimal solution as intended particle prediction
Position, i.e., be concentrated through calculating the final export structure of method choice of expected utility from more excellent solution.
The present invention is described in detail above in association with accompanying drawing embodiment, those skilled in the art can be according to upper
State it is bright the present invention is made many variations example.Thus, some of embodiment details should not constitute limitation of the invention, this
Invention regard the scope defined using appended claims as protection scope of the present invention.
Claims (6)
1. a kind of Advances in protein structure prediction based on multi-objective particle swarm optimization, including step：
S1：Selection needs n optimized different initial configurations as n intended particle in a protein sequence to be predicted,
And represent the position coordinates of each intended particle in the form of conformation is represented, n is the natural number more than zero；
S2：Parameter setting is carried out to a more new formula；
S3：Three target energy formula are iterated according to the more new formula, and obtain a more excellent disaggregation；
S4：Handle the optimal solution set and obtain an optimal solution, and using the optimal solution as the intended particle predicted position.
2. the Advances in protein structure prediction based on multi-objective particle swarm optimization according to right wants 1, it is characterised in that institute
Stating more new formula includes formula (1) and formula (2)：
Wherein, w is inertia coeffeicent, and k is iterations, c1For cognitive coefficient, c2For coefficient of association；It is described for renewal speed
Renewal speed represents that the i-th particle compares the speed for changing structure in upper once iterative process in kth time iterative process；Rand is
Random number between 0~1；For history optimum structure；For global optimum's structure；It is the i-th particle
The position coordinates in k iterative process.
3. the Advances in protein structure prediction based on multi-objective particle swarm optimization according to right wants 2, it is characterised in that institute
State in S2 steps, the renewal speed for initializing each intended particle is 0, and to a maximum iteration, described used
Property coefficient, the cognitive coefficient and the coefficient of association are configured.
4. the Advances in protein structure prediction based on multi-objective particle swarm optimization according to right wants 3, it is characterised in that institute
Stating target energy formula includes Rosetta energy functions, QUARK energy functions and CHARMM energy functions.
5. the Advances in protein structure prediction based on multi-objective particle swarm optimization according to right wants 3 or 4, its feature exists
In the S3 steps further comprise step：
S31：It is utilized respectively the target energy formula and calculates one first energy function value for obtaining each primary, one the
Two energy function values and one the 3rd energy function value；
S32：It is determined that and select non-dominant particle in each intended particle, and the non-dominant particle is added into a more excellent solution
Collection；The first energy function value corresponding to the non-dominant particle is the minimum value in each first energy function value, described
The second energy function value corresponding to non-dominant particle is the minimum value in each second energy function value, the non-dominant grain
The 3rd energy function value corresponding to son is the minimum value in each 3rd energy function value, and first energy function value
It is unique minimum value in each first energy function value or second energy function value is each second energy function
Unique minimum value or the 3rd energy function value are unique minimum value in each 3rd energy function value in value；
S33：The renewal speed according to corresponding to the formula (1) updates each intended particle；
S34：The position coordinates of each intended particle is updated according to the formula (2)；
S35：It is utilized respectively target energy formula described in three and calculates the first energy function described in the one of each intended particle of acquisition
Value, the 3rd energy function value described in the second energy function value and one described in one；
S36：It is determined that and select non-dominant particle in each presently described intended particle, and by the non-dominant particle add described in more
Excellent disaggregation；
S37：Judge whether iterations reaches the maximum iteration；Such as no, return to step S33；In this way, follow-up step is continued
Suddenly.
6. the Advances in protein structure prediction based on multi-objective particle swarm optimization according to right wants 5, it is characterised in that institute
State S4 and further comprise step：
S41：Build a utility function, U=λ1f1+λ2f2+λ3f3, wherein f1For first energy function value, f2For described second
Energy function value, f3For the 3rd energy function value, λ1For the first weight, λ2For the second weight, λ3For the 3rd weight, λ1、λ2
And λ3Span be respectively [0,1] and λ1+λ2+λ3=1；
S42：To λ1、λ2N times stochastical sampling is carried out enough, and the calculating more excellent solution concentrates the phase corresponding to each non-dominant particle
Hope effectiveness E (U)；
Choose numerical value it is maximum one described in corresponding to expected utility one described in non-dominant particle be used as affiliated optimal solution.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710244181.6A CN107025383A (en) | 2017-04-14 | 2017-04-14 | Advances in protein structure prediction based on multi-objective particle swarm optimization |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201710244181.6A CN107025383A (en) | 2017-04-14 | 2017-04-14 | Advances in protein structure prediction based on multi-objective particle swarm optimization |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN107025383A true CN107025383A (en) | 2017-08-08 |
Family
ID=59527334
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201710244181.6A Pending CN107025383A (en) | 2017-04-14 | 2017-04-14 | Advances in protein structure prediction based on multi-objective particle swarm optimization |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN107025383A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107704726A (en) * | 2017-10-13 | 2018-02-16 | 齐齐哈尔大学 | A kind of method that Protein folding prediction is carried out using population simulated annealing |
| CN110610742A (en) * | 2019-09-20 | 2019-12-24 | 福建工程学院 | Functional module detection method based on protein interaction network |
| CN111554346A (en) * | 2020-04-29 | 2020-08-18 | 上海交通大学 | Protein sequence design implementation method based on multi-objective optimization |
| CN112086132A (en) * | 2020-08-18 | 2020-12-15 | 深圳晶泰科技有限公司 | Organic molecular crystal construction method and system |
| CN113035268A (en) * | 2021-04-09 | 2021-06-25 | 上海交通大学 | Protein structure optimization method based on multi-objective decomposition optimization strategy |
| WO2022006771A1 (en) * | 2020-07-08 | 2022-01-13 | 深圳晶泰科技有限公司 | Molecular force field multi-objective fitting algorithm database system and workflow method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090089035A1 (en) * | 2007-07-07 | 2009-04-02 | Solomon Research Llc | Hybrid multi-layer artificial immune system |
| CN103886225A (en) * | 2012-12-21 | 2014-06-25 | 中国科学院大连化学物理研究所 | Method for designing proteins on basis of polarizable force fields and pso (particle swarm optimization) |
| CN105930688A (en) * | 2016-04-18 | 2016-09-07 | 福州大学 | Improved PSO algorithm based protein function module detection method |
| WO2017011779A1 (en) * | 2015-07-16 | 2017-01-19 | Dnastar, Inc. | Protein structure prediction system |
-
2017
- 2017-04-14 CN CN201710244181.6A patent/CN107025383A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20090089035A1 (en) * | 2007-07-07 | 2009-04-02 | Solomon Research Llc | Hybrid multi-layer artificial immune system |
| CN103886225A (en) * | 2012-12-21 | 2014-06-25 | 中国科学院大连化学物理研究所 | Method for designing proteins on basis of polarizable force fields and pso (particle swarm optimization) |
| WO2017011779A1 (en) * | 2015-07-16 | 2017-01-19 | Dnastar, Inc. | Protein structure prediction system |
| CN105930688A (en) * | 2016-04-18 | 2016-09-07 | 福州大学 | Improved PSO algorithm based protein function module detection method |
Non-Patent Citations (2)
| Title |
|---|
| 刘裕峰: ""统计势能函数及其在蛋白质结构预测中的应用"", 《中国博士学位论文全文数据库 基础科技辑》 * |
| 焉为家: ""利用改进的粒子群算法预测蛋白质结构"", 《中国优秀硕士学位论文全文数据库 基础科技辑》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107704726A (en) * | 2017-10-13 | 2018-02-16 | 齐齐哈尔大学 | A kind of method that Protein folding prediction is carried out using population simulated annealing |
| CN110610742A (en) * | 2019-09-20 | 2019-12-24 | 福建工程学院 | Functional module detection method based on protein interaction network |
| CN110610742B (en) * | 2019-09-20 | 2023-12-19 | 福建工程学院 | Functional module detection method based on protein interaction network |
| CN111554346A (en) * | 2020-04-29 | 2020-08-18 | 上海交通大学 | Protein sequence design implementation method based on multi-objective optimization |
| CN111554346B (en) * | 2020-04-29 | 2023-05-23 | 上海交通大学 | Protein sequence design implementation method based on multi-objective optimization |
| WO2022006771A1 (en) * | 2020-07-08 | 2022-01-13 | 深圳晶泰科技有限公司 | Molecular force field multi-objective fitting algorithm database system and workflow method |
| CN112086132A (en) * | 2020-08-18 | 2020-12-15 | 深圳晶泰科技有限公司 | Organic molecular crystal construction method and system |
| CN113035268A (en) * | 2021-04-09 | 2021-06-25 | 上海交通大学 | Protein structure optimization method based on multi-objective decomposition optimization strategy |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN107025383A (en) | Advances in protein structure prediction based on multi-objective particle swarm optimization | |
| CN106778059B (en) | A kind of group's Advances in protein structure prediction based on Rosetta local enhancement | |
| CN103208027B (en) | Method for genetic algorithm with local modularity for community detecting | |
| Lyubartsev et al. | Systematic hierarchical coarse-graining with the inverse Monte Carlo method | |
| Wu et al. | The exploration/exploitation tradeoff in whale optimization algorithm | |
| CN103455610B (en) | Network community detecting method based on multi-objective memetic computation | |
| CN106650305B (en) | A kind of more tactful group Advances in protein structure prediction based on local abstract convex supporting surface | |
| CN105260786A (en) | Comprehensive optimization method of simulation credibility evaluation model of electric propulsion system | |
| Zhao et al. | Identifying N6-methyladenosine sites using extreme gradient boosting system optimized by particle swarm optimizer | |
| Zamani et al. | Amino acid encoding schemes for machine learning methods | |
| CN108427865A (en) | A method of prediction LncRNA and environmental factor incidence relation | |
| Hoffgaard et al. | BioPhysConnectoR: Connecting sequence information and biophysical models | |
| Tümer et al. | Application of radial basis function networks with feature selection for gdp per capita estimation based on academic parameters | |
| Zhou et al. | Enhanced hybrid search algorithm for protein structure prediction using the 3D-HP lattice model | |
| CN103473599A (en) | Genetic algorithm and Kalman filtering based RBFN (Radial Basis Function Networks) combined training method | |
| Preuss et al. | Solving phase equilibrium problems by means of avoidance-based multiobjectivization | |
| CN107358001A (en) | A kind of constrained global optimization method based on Kriging models | |
| Zhang et al. | Predicting protein-ligand binding affinity via joint global-local interaction modeling | |
| Cui et al. | Swarm intelligence in bioinformatics: Methods and implementations for discovering patterns of multiple sequences | |
| Giscard et al. | Cycle-centrality in economic and biological networks | |
| Sun et al. | A multiscale model to predict the elastic property of microtubules | |
| Arkin | Comparison of the Parallel Tempering algorithm and multicanonical method as applied to coarse-grained off-lattice models for folding heteropolymers | |
| Liu et al. | An innovative function-tree building method based on similarity theory and extension theory | |
| Ma et al. | Predicting protein-protein interactions based on BP neural network | |
| Zhou et al. | An artificial neural network method for combining gene prediction based on equitable weights |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20170808 |
|
| RJ01 | Rejection of invention patent application after publication |